Abstract In this study, we will identify somatic tumor mutations, CNVs and a melanoma immune profile in relationship to survival in order to identify patients with poor prognosis among AJCC TNM Stages IIA-IIIB and eventually distinguish which patients may profit from adjuvant therapies and save lives. Our hypothesis is that primary melanomas will have molecular and clinical features that will allow the prognostic stratification of melanoma tumors among these patients. The current survival rate of individuals diagnosed at these stages ranges from 22 ? 82%. There is little understanding of which patients will progress and which will not based on stage. The ultimate intent of this project is to provide information to translate to the clinic to personalize care. Currently, there is a dearth of studies in the melanoma field looking at epidemiology/genomic factors and melanoma survival in order to identify, confirm and develop such biomarkers. We have brought together nine cohorts of melanoma patients of 1,000 individuals and their tumors, half who have died from melanoma within five years (median survival, 2.4 years) and half who have lived for at least five years (median follow up 8.5 years). Patients will be frequency matched for stage. Data will be randomly divided into a training set of 660 tumors and a validation set of 340 tumors, equally divided by poor and good prognosis. All patients have been treated using standard-of-care surgery; they have adequate tumor tissue, germline DNA and clinical, pathologic and demographic information recorded. Project 1 will use targeted next generation sequencing of clinically actionable mutations and CNV?s in order to develop subgroups of patients. Secondly, we will evaluate a melanoma immune profile, CD3/CD8/FOXP3 slides stained with immunofluorescence and PD1-PDL1 axis chromogenic staining, to compare to a strong pathologic variable associated with survival, tumor infiltrating lymphoctyes (TILs). We will also evaluate ratios of CD8:FOXP3 and CD8:PD-L1 as prognostic. The melanoma immune profile may be more quantitative and reproducible than TIL score and we will determine this in this large study. Finally, our strong multidisciplinary team of clinicians, biostatisticians, laboratory scientists and epidemiologists will help to develop new strategies and new information to understand molecular factors associated with prognosis in melanoma. Specifically, it will help to identify which individuals are likely to have a poor prognosis and potentially identify these for new adjuvant therapy and closer surveillance.